Herbert Markman and Positek, Inc. v. Westview Instruments, Inc. And Althon Enterprises, Inc.

PAULINE NEWMAN, Circuit Judge,

dissenting.

I

INTRODUCTION

The issue is the role of the jury in patent infringement cases. The majority opinion resolves the issue by designating, as law, factual disputes about the meaning and scope of the technologic terms and words of art used to define patented inventions. By holding that these disputed technologic questions are matters of law, the court holds that issues of patent infringement, previously triable to a jury as of right, will now be decided by the trial judge and then re-decided de novo by this court on appeal.

Patent infringement is a factual question. Its resolution often requires finding the factual meaning and scope of the terms of scientific art and technology and usage by which the patentee described and claimed the invention. These findings usually require testamentary and documentary evidence and occasionally experiments or demonstrations, as illustrated in many of our previous decisions that are now overruled.

Deciding the meaning of the words used in the patent is often dispositive of the question of infringement. Thus in the case at bar the infringement controversy is decided by finding the meaning and scope of the term “inventory” in Markman’s patent, in light of the accused Westview system: if “inventory 5 ’ is limited to clothing, the patent is not infringed; if “inventory” includes invoices, it is. The majority holds that this is a matter of law, devoid of any factual component; and subject to de novo appellate determination. The jury is eliminated, and new and uncertain procedures are imposed on trial judges.

This holding not only raises a constitutional issue of grave consequence, but the court creates a litigation system that is unique to patent cases, unworkable, and ultimately unjust. Thus I must, respectfully, dissent.

I shall discuss three principal concerns:

1. The Meaning of Disputed Technologic Terms and Words of Art

Patents are technologic disclosures, written by and for the technologically experienced: those “of skill in the art.” The meaning and scope of the terms that define the patented technology is often in dispute in infringement litigation, for it often decides the case. In resolving such dispute the trier of fact often makes findings that depend on the weight, credibility, and probative value of conflicting evidence, such as that offered on behalf of Markman and Westview. Heretofore, the disputed meaning of technologic terms and words of art has been treated by Federal Circuit precedent as an “underlying fact” on which the legal effect of the patent is based. The majority now simply rules that these are not “underlying factual inquiries.” However, the meaning and scope of disputed technologic and other terms of art in particular usage are classical questions of fact. Their nature as fact does not change because their finding, like most findings in litigation, has a legal consequence. By redesignating fact as “law” the court has eliminated the jury right from most trials of patent infringement.

2. The Trial and Appellate Roles in Tech-nologic Disputes

The trial process is the vehicle for determining truth. Thus the trier of fact is present in the courtroom along with the witnesses, the advocates, the exhibits, and the demonstrations. Indeed, when the techno-logic issues are complex, appellate fact finding is probably the least effective path to accurate decisionmaking. And if a factual question is technologically simple, it is not thereby transformed into a matter of law and removed from the trial process. Even were there no constitutional infirmity, I doubt that the correct resolution of technologic or scientific disputes is more likely to be achieved by removing disputed facts from the procedures of trial and consigning them to the appellate court. Appellate briefs and fifteen minutes per side of attorney argument are not designed for de novo findings of disputed tech-nologic questions.

*1000

3. The Constitution

Jury trial in patent cases is protected by the Seventh Amendment. Elimination of the jury is not this court’s choice to make.

The constitutional right alone bars the majority’s new rule. The majority today denies 200 years of jury trial of patent cases in the United States, preceded by over 150 years of jury trial of patent cases in England, by simply calling a question of fact a question of law. The Seventh Amendment is not so readily circumvented.

II

FACT AND LAW IN PATENT INFRINGEMENT

A LEGAL “CONSTRUCTION” v. FACTUAL “INTERPRETATION”

The majority’s explanation for removing these factual issues from the jury is that it is “construing” the patent claims, and that the “construction” of documents is a matter of law. The legal construction of documents— patent documents and other documents — is indeed a matter of law. The legal effect of the patent claim is to establish the metes and bounds of the patent right to exclude; this is a matter of law. But this does not deprive the underlying facts of their nature as fact. These facts are found on evidence that includes the patent specification, relevant prior art, the prosecution history, the testimony of experts in the field, and other relevant evidence such as tests and demonstrations, all as I shall discuss post. These findings do not become rules of law because they relate to a document whose legal effect follows from the found facts.

An extensive body of law, statutory and judgemade, governs the construction and legal effect of patent claims; for example, that a claim is construed the same way in determining both patent validity and infringement; that a dependent claim includes all of the limitations of the independent claim; that the claims as filed are part of the technical disclosure; that the right to exclude is divisible into making, using, or selling the claimed subject matter; that a claim is not infringed unless every element thereof is met in the accused device, either literally or by an equivalent. These and other rules of law are applied when appropriate to the facts of the particular ease: either undisputed facts, or facts that are found by the trier of fact. The procedure of applying law to facts does not convert the finding of facts into a matter of law.

In patent infringement litigation there is often a factual dispute as to the meaning and scope of the technical terms or words of art as they are used in the particular patented invention. When such dispute arises its resolution is not a ruling' of law, but a finding of fact. Such findings of meaning, scope, and usage have been called the “interpretation” of disputed terms of a document, as contrasted with the “construction” or legal effect of a document. Professor Corbin has explained this distinction, in the context of contracts, as reflecting the difference between “language” and the “legal operation” of language:

It may be helpful to note that the word interpretation is commonly used with respect to language itself — to the symbols (the words and acts) of expression. In about the same degree, we speak of the construction of a contract. It is true that we also speak of construing language and of interpreting a contract; but by the latter phrase is certainly meant interpreting the words of a contract. The word “contract” has been variously defined; but it is seldom identified with mere symbols of expression. By “interpretation of language” we determine what ideas that language induces in other persons. By “construction of the contract,” as that term will be used here, we determine its legal operation — its effect upon the action of courts and administrative officials.

3 Arthur L. Corbin, Corbin on Contracts § 534 (1960) (footnotes omitted). The Restatement (Second) of Contracts § 200 and Comment c (1981) describes the distinction between “construction” and “interpretation” as reflecting the difference between the “meaning” of a term and its “legal effect”:

Interpretation of a promise or agreement or a term thereof is the ascertainment of its meaning.

* * * * * *

*1001 Interpretation is not a determination of the legal effect of words or other conduct.

The Reporter’s Note explains that the purpose is “to make it clear that ‘interpretation’ relates to meaning and to avoid confusion with the ascertainment of legal operation or effect, sometimes called ‘construction.’ ” (Citations omitted.) The analogy is apt, although a patent is not a contract, for this distinction has been recognized for many kinds of written instruments. See, e.g., In re XTI Xonix Technologies Inc., 156 B.R. 821 , 829 n. 6 (D.Ore.1993) (proceeding in bankruptcy):

Interpretation and construction of written instruments are not the same. A rule of construction is one which either governs the effect of an ascertained intention, or points out what the court should do in the absence of express or implied intention, while a rule of interpretation is one which governs the ascertainment of the meaning of the maker of the instrument.

Williams v. Humble Oil & Ref. Co., 432 F.2d 165 , 179 (5th Cir.1970), cert. denied, 402 U.S. 934 , 91 S.Ct. 1526 , 28 L.Ed.2d 868 (1971) (contract):

In the law of contracts (conventional obligations) a proper distinction exists between the “interpretation” of written instruments and their “construction.” “Interpretation” refers to the process of determining the meaning of the words used; that process is traditionally thought to be a function of the jury. On the other hand, the process of determining the legal effect of the words used — once we know their meaning — is properly labelled “construction”; it is peculiarly a function of the court.

Hornick v. Owners Ins. Co., 511 N.W.2d 370 (Iowa 1993) (insurance policy):

Construction of an insurance, policy — the process of determining its legal effect — is a question of law for the court. Interpretation — the process of determining the meaning of words used — is also a question of law for the court unless it depends on extrinsic evidence or a choice among reasonable inferences to be drawn.

In re Union Trust Co., 89 Misc. 69 , 151 N.Y.S. 246, 249-50 (Sur.Ct.1915) (will):

A rule of construction is one which either governs the effect of an ascertained intention or points out what a court should do in the absence of express or implied intention. A rule of interpretation is one which governs the ascertainment of the meaning of the maker of a written document.

Reed v. Proprietors of Locks & Canals on Merrimac River, 49 U.S. (8 How.) 274, 288-89 , 12 L.Ed. 1077 (1850) (deed):

It is true, that it was the duty of the court to give a construction to the deed in question, so far as the intention of the parties could be elicited therefrom.... But after all this is done, it is still a question of fact to be discovered from evidence dehors the deed ... for the jury to decide, whether the land in controversy is included therein, or, in other words, was intended by the parties so to be.

It is indeed well understood that the legal effect or construction of the terms of a document, a matter of law, is not to be confused with resolution of disputes concerning the factual meaning of the terms. The former is for the court, the latter for the jury. That the thing whose terms require interpretation is a patent, instead of a deed or a will or a contract, does not convert the finding of disputed facts into a matter of law. Factual findings concerning a particular patented invention do not become matters of law simply because the patent document serves a legal purpose.

Although purity of language has occasionally slipped, for the words “construction” and “interpretation” have been loosely used, the distinction between the concepts has been recognized when it mattered. For example, Walker in his 1904 Textbook used the phrase “construction of the patent,” but he left no doubt as .to the role of the jury as trier of fact:

[Wjhere the question of infringement depends on the construction of the patent, and that construction depends upon a doubtful question in the prior art, the latter question should be left for the jury; and the dependent question of infringement should also be left for the jury to decide.

*1002 A.H. Walker, Textbook on the Patent Laws of the United States of America § 536 (4th ed. 1904).

This recognition that the factual issues that underlie the “construction of the patent,” and that determine patent infringement, are for the jury is manifest even in the early Supreme Court cases that are relied on by the majority, as I discuss in Part IV-C, post. The majority’s authority does not show removal' of factual disputes from the jury. Indeed, several of the cases that are relied upon were bills in equity, and irrelevant to jury trials.

B. EVIDENCE RELEVANT TO CLAIM INTERPRETATION

The areas of evidentiary inquiry commonly encountered in patent infringement cases are illustrated in Markman’s case. The infringement trial (validity and damages were severed) included evidence relating to the patent specification, the patent claims, the prosecution history, the inventor’s usage of “inventory,” and the defendant’s understanding of the term. Although Markman’s invéntion was not complex, this phase of the trial took three days.

1. The Specification

The patent specification contains the description of the invention, including the claims. It fulfills the inventor’s obligation to make known the technology for which the patent is granted, and must meet certain legal requirements. It must be written; and clear. It must be complete yet concise. It must enable one of skill in the field to make and use the invention, but need not include that which is known to the field. It must describe the best mode known to the inventor. It is a technical document, written for persons experienced in the technology. See 2 Irving Kayton et al., Patent Practice 9-1 to 9-3 (4th ed. 1989) (describing thirteen functions of the patent specification).

The claims are part of the specification. Their purpose is to identify — “particularly pointing out and distinctly claiming,” 35 U.S.C. § 112 — that which is the subject of the patent grant. Patent claims are terse summaries, and do not repeat the technologic content of the specification. When there arises a question as to whether a term in a patent claim is of a meaning and scope that reaches particular subject matter, the interested public and in turn the courts look to the body of the specification for elaboration and illustration of the usage of the term to define the patented technology.

When litigation ensues, it may be helpful to the trier of fact to hear from the inventor what he/she meant by the terms of art and science and technology used to describe the invention. Markman, as the inventor, testified on this aspect. It may also be helpful to hear, from others in the field of the invention, what was conveyed to them by the now-disputed terms. Markman’s witness and Westview’s president so testified. Not unexpectedly, the evidence was conflicting. The trier of fact may have to assess the technical content, weight, and credibility of all of the evidence, including but not limited to the specification, in finding the meaning and scope of disputed terms as used in the patent. Whether the term “inventory” in Mark-man’s Claim 1 includes Westview’s invoices can not be found in the abstract, nor by consulting a dictionary; it is found on the evidence of this case, for the specific invention and the specific accused system. This has historically been a question of fact. 1

The Federal Circuit has explained the rela- • tionship between law and fact in claim interpretation in many cases, all now overruled, as I illustrate in Part IV-A, B, post. For example, in Perini America, Inc. v. Paper Converting Mach. Co., 832 F.2d 581 , 4 USPQ2d 1621 (Fed.Cir.1987) the patented invention related to machines having embosser rolls *1003 used in manufacturing paper towels. Typical of the technologic terms in dispute was “the projections in one web intermediate the projections in the other web” to describe the alignment of the embosser rolls. This court recognized that the meaning of this term was a factual issue, and explained that claim interpretation rests on underlying facts:

Like all legal conclusions, that conclusion [that a claim must be interpreted in a certain way] rises out of and rests on a foundation built of established (undisputed or correctly found) facts. Interpretation of a claim, or of its scope, should not be assayed until a foundation is in place. If the meaning of terms in the claim, the specification, other claims, or prosecution history is disputed, that dispute must be resolved as a question of fact before interpretation can begin.

Id. at 584 , 4 USPQ2d at 1624. The court observed that the finding of disputed facts can “dictate” the ultimate conclusion of what the claim means:

Confusion may be caused by the circumstance in which resolution of the question of the meaning of a term or terms dictates the interpretation of the claim, but that is not unusual, legal conclusions being dictated by established facts and not the other way around, and does not change the nature of the meaning-of-terms inquiry from one of fact to one of law.

Id. The majority now criticizes and expressly overrules these statements in Perini, holding that these facts are not fact, but “law,” and that they are removed from the trier of fact and are determined de novo on appeal. In my view the Perini court’s analysis is in accord with precedent, and properly preserves the role of the trier of fact, whether trial is to the bench or to a jury.

2. Prior Art

When there is a dispute as to the meaning and scope of a technologic term or word of art in a patent claim, it is often helpful to look at the prior art: that is, what was known to persons in the field of the invention at the time the invention was made. The scope and content of the prior art, the differences between the claimed invention and the prior art, the level of ordinary skill in the field of the invention, are all questions of fact, Graham v. John Deere Co., 383 U.S. 1, 17 , 86 S.Ct. 684, 693 , 15 L.Ed.2d 545 , 148 USPQ 459, 467 (1966), and are not subject to reclassification by us. 2

The prior art may provide evidence of how the disputed technologic terms and words of art or science were used by others in the field of the invention, and thus evidence of what was conveyed to the field by the terms as used by the patentee. Indeed, the infringement analysis can sometimes stop with the prior art, for if the accused device is found in the prior art; then it is a rule of law that the patent claims can not be interpreted to reach that device. This too requires findings of the scope and content of the prior art, a question of fact, Graham v. John Deere, that is found with an eye upon the accused device.

I do not attempt to catalogue the myriad kinds of information, findings, and inferences that may flow from the prior art, in determining the technologic scope of the paten-tee’s invention. This evidence, and the findings and inferences that are drawn, are the province of the trier of fact. The majority’s insistence that these are purely legal matters of “claim construction” does indeed serve to replace the trier of fact with the Federal Circuit; I doubt that it improves the quality of the decision, at great cost to efficiency of the trial/appellate process.

3. The Prosecution History

The prosecution history is the record in the Patent and Trademark Office of what transpired during examination of the patent application. It is a public record. It sometimes is lengthy and detailed, sometimes *1004 sketchy and brief. The prosecution history may provide evidence of how the inventor or the patent examiner viewed the now-disputed technologic and other terms. In Howes v. Medical Components, Inc., 814 F.2d 638, 645 , 2 USPQ2d 1271, 1274-75 (Fed.Cir.1987) this court observed that “during the prosecution of a patent, claim language may take on new meanings, possibly different from that which was originally intended.” The way patenta-bility was argued by the inventor, concessions made or positions adjusted, may be relevant to the factual issue in dispute, and may create an estoppel against the patentee’s now-proposed interpretation.

The determination of what occurred in the prosecution of the patent application is a factual matter, Smithkline Diagnostics Inc. v. Helena Lab. Corp., 859 F.2d 878, 882 , 8 USPQ2d 1468, 1471-72 (Fed.Cir.1988), specific to the particular patent. It is often based on technological arguments, experimental evidence submitted to the patent office, discussions of the meaning and relevance of prior publications and prior knowledge, explanations of the technical content of the specification, and other evidence of the applicant’s and the examiner’s positions. This evidence, and appropriate findings and inferences, are for the trier of fact.

The majority refers to the “undisputed” prosecution history, in asserting that there are no factual aspects to this evidence. Indeed, the official government record is fixed. But the significance of the exchanges, compromises, and explanations contained in the correspondence between the inventor and the examiner; the inferences to be drawn as to the technology, the invention, and the meaning and scope of now-disputed technologic terms or words of art; may depend on this and other evidence. If disputed, their finding is for the trier of fact. The meaning, significance, and weight of the content of a documentary record does not become a matter of law simply because the content of the record is not in dispute.

4. Technologic/Scientific Facts

Decision of the question of patent infringement usually turns on findings of technologic fact; sometimes relatively simple technology, sometimes at the frontier of scientific advance and its practical applications. When scientific and technologic disputes arise in litigation, they are subject to the rules of evidence and procedure. See Daubert v. Merrell Dow Pharmaceuticals, Inc., — U.S. -, 113 S.Ct. 2786 , 125 L.Ed.2d 469 (1993) (discussing issues of scientific evidence). The complexity of the technologic/scientific evidence will of course vary with the issue. See, e.g., Brooktree Corp. v. Advanced Micro Devices, Inc., 977 F.2d 1555 , 24 USPQ2d 1401 (Fed.Cir.1992) (color video display semiconductor chips); Scripps Clinic & Research Found, v. Genentech, Inc., 927 F.2d 1565 , 18 USPQ2d 1001 (Fed.Cir.1991) (blood clotting factor VIII); Halliburton Co. v. Schlumberger Technology Corp., 925 F.2d 1435 , 17 USPQ2d 1834 (Fed.Cir.1991) (neutron logging of oil wells); Bey v. Kollonitsch, 806 F.2d 1024 , 231 USPQ 967 (Fed.Cir.1986) (irreversible enzyme inhibitors); Fromson v. Advance Offset Plate, Inc., 755 F.2d 1549 , 225 USPQ 26 (Fed.Cir.1985) (silicated lithography plates).

The Court stressed in Daubert that the admissibility of scientific evidence depends on its reliability and relevance, and that the judge’s responsibility, when the trier of fact is a jury, is to assure the adequacy of the methodology upon which the evidence is based. — U.S. at -, 113 S.Ct. at 2795-96 . This emphasis on methodology is as well suited to practical applications of technology and engineering as to basic scientific principles. Evidence in patent cases is often provided by scientists or engineers as expert witnesses, and may include explanations of the technology and its scientific basis, comparisons with the prior art or with the accused device, experiments, demonstrations, and interpretations. The evaluation of technologic evidence is often required of the trier of fact. See Lee Loevinger, Science as Evidence, 35 Jurimetrics J. 153 (1995); Jack B. Weinstein, The Effect of Daubert on the Work of Federal Trial Judges, 2 Shepard’s Expert and Scientific Evidence Quarterly 1 (1994).

Nor is it rare in patent cases to encounter incomplete data, theoretical uncertainties, untested inferences, and speculative conclu *1005 sions. Experimental procedures, the sources of data, and the bases of opinions that are offered to prove/disprove a technologic fact are often in evidentiary conflict in patent disputes. Engineers and scientists know very well the uncertainties of the experimental process, the fluctuations and glitches in the data, the human and machine error, the forks in the road to objective truth. Indeed, understanding of the fallibility of technologic and scientific experimentation is soon acquired by those who labor in the field of litigation. “The community of trial lawyers and judges knows perhaps better than any other professional group just how unruly science often is in practice.” Sheila Jasanoff, What Judges Should Know About the Sociology of Science, 11 Judicature 77, 80 (1993) (discussing the “social dimension [that] gives legitimacy to particular scientific ‘facts’”).

Now that the Federal Circuit holds that resolution of disputes as to the meaning and scope of technologic terms and words of art as used in a particular patent is law, not fact, removing the jury from this issue, is the trial judge excused from determining the admissibility and relevance of technologic evidence? What about the requirement that “Credibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of a judge.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255, 106 S.Ct. 2505, 2513 , 91 L.Ed.2d 202 (1986). In a patent case the trier of fact may receive extensive evidence related to the meaning and scope of technologic or scientific terms or words of art, their usage and their perception in the field of a particular invention. The evidence often includes technical publications, scientific articles, experimental data, demonstrations, and opinion testimony. The factual nature of such evidence can not be squared with the majority’s criticism of such eases as Palumbo v. Don-Joy, 762 F.2d 969 , 226 USPQ 5 (Fed.Cir.1985), which is today overruled for its holding that “when the meaning of a term in a claim is disputed and extrinsic evidence is necessary to explain that term, then an underlying factual question arises.” Id. at 974 , 226 USPQ at 8 . In addition to my concern about how a record will be developed for the Federal Circuit’s de novo decision, I doubt that an appellate court’s de novo finding of technologic facts is more likely to attain accuracy, than the decision of a jury or judge before whom a full trial was had:

Duplication of the trial judge’s efforts in the court of appeals would very likely contribute only negligibly to the accuracy of fact determination at a huge cost in diversion of judicial resources.

Anderson v. City of Bessemer City, 470 U.S. 564, 574-75 , 105 S.Ct. 1504, 1512 , 84 L.Ed.2d 518 (1985).

5. The Testimony of Experts

Disputed questions of the meaning and scope of technologic terms and words of art are decided from the viewpoint of persons of skill in the specific field of technology. It is rare to come upon a' technologic issue in litigation for which differing and often plausible views are not offered by qualified witnesses. The Federal Rules of Evidence contemplate the provision of specialized knowledge to assist the trier of fact:

R. 702. If scientific, technical,' or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto_

In Moeller v. Ionetics, Inc., 794 F.2d 653 , 229 USPQ 992 (Fed.Cir.1986) this court' held that the trial court’s rejection of expert testimony in that case was an abuse of discretion:

Although use of experts is generally a matter of discretion with the trial judge, that discretion is not unlimited. In a patent case involving complex scientific principles, it is particularly helpful to see how those skilled in the art would interpret the claim.

Id. at 657 , 229 USPQ at 995 (citations omitted). In Moeller the claims related to an electronic process for measuring the concentration of cations, the dispute centering on the particular meaning of the term “electrode body.” Moeller too is now criticized and overruled for its statement that

although claim construction is a legal question, underlying fact disputes may arise *1006 pertaining to extrinsic evidence that might preclude summary judgment treatment of claim construction.

Id. (citation omitted). I do not see how the Federal Circuit could have decided, de novo on appeal, the meaning of “electrode body” in this particular invention without finding disputed technologic facts.

The majority’s stated recognition that expert testimony may be useful, while holding that “extrinsic evidence of record cannot be relied on to change the meaning of the claims,” majority op. at 998, denying all deference to the trier of fact’s findings based on that evidence, illustrate the confusion in the court’s plan of de novo claim interpretation as a matter of law. The majority has created a procedural quandary, for extrinsic evidence can apparently be received, but no jury can weigh it. When the extrinsic evidence is in conflict — as it invariably is — what then? Will the Federal Circuit itself weigh the evidence of expert witnesses? Will we receive a collection of self-serving affidavits, without examination and cross-examination? Such a procedure surely is not optimal for cases that may require decision of complex engineering or electronics, or chemical or biological processes.

The Court in Daubert referred to the value of the adversary system in matters of scientific proof. The cross-examination of technical experts, with the adversarial guidance of other technical experts, can be as rigorous as any “peer review” process. See generally Margaret A. Berger, Procedural Paradigms for Applying the Daubert Test, 78 Minn. L.Rev. 1345 (1994). In resolving litigation controversy by determining mechanical or chemical or electronic truth, it is hard to understand why justice should be handicapped in the Federal Circuit by replacement of a live trial with cold documents.

In eliminating all sources of “fact” that might implicate the jury right, the majority has denied to the trial of patent cases the assistance that Federal Rule of Evidence 702 is designed to provide, as well as the benefits of Rules 403, 703, and 706. The purpose of these rules is “that the truth may be ascertained and proceedings justly determined.” Fed.R.Evid. 102. It seems to me that we have constructed a Hobson’s choice whereby either (1) there will indeed be factual evidence of technologic meaning entered into the trial record, for de novo decision on the record by the Federal Circuit, (2) there will be scant evidence admitted at trial, in view of our pronouncement that there is only law in claim interpretation. Either way, one might call this the “omniscience of the learned man” theory of dispute resolution in the Federal Circuit. 3

Findings of the meaning of technologic terms and words of art in particular usages are the province of the trier of fact. Discussing words and their jurisprudential treatment, Justice Holmes wrote:

A word is not a crystal, transparent and unchanged; it is the skin of a living thought and may vary greatly in color and content according to the circumstances and the time in which it is used.

Towne v. Eisner, 245 U.S. 418, 425 , 38 S.Ct. 158, 159 , 62 L.Ed. 372 (1918). In Autogiro Co. of America v. United States, 384 F.2d 391, 397 , 181 Ct.Cl. 55 , 155 USPQ 697, 702 (1967) one of our predecessor courts remarked: “The very nature of words would make a clear and unambiguous [patent] claim a rare occurrence.” Justice Story explained the roles of judge and jury with respect to the meaning of “words of art, and technical phrases” in patent documents:

In respect to another objection, viz. that the court was bound to state what in point of law the invention claimed by the paten-tee was, I agree, that this is generally true, so far as the construction of the words of the patent, and specification is concerned. But then this doctrine is to be received with qualifications, and sub modo, as the very opinion of Mr. Baron Parke, cited by the counsel, in the case of Neilson v. Har-ford, Webster Pat.Cas. 295, 370,[ 4 ] abundantly shows; and the jury are to judge of *1007 the meaning of words of art, and technical phrases, in commerce and manufactures, and of the surrounding circumstances, which may materially affect, enlarge or control the meaning of the words of the patent and specification.

Washburn v. Gould, 29 F.Cas. 312, 325 (C.C.D.Mass.1844) (emphasis added). Justice Story recognized that the meaning of words of art may depend on “the surrounding circumstances.” Indeed, the Federal Circuit recognized that words do not always have the same meaning when they are adapted to new uses. See Fromson v. Advance Offset Plate, Inc., 720 F.2d 1565, 1569 , 219 USPQ 1137, 1140 (Fed.Cir.1983) (patentee may be his own lexicographer).

Inventors’ usages of words to describe their inventions, and the meaning thereby conveyed to persons skilled in the field, are questions of fact, not matters of law, in patent documents as in other written instruments. Disputes concerning the meaning and usage of technical terms and words of art arise in many areas of law. These disputes are resolved by the triers of fact, whether judge or jury, in their established roles in the adjudicatory process. For example, the role of the jury with respect to technical terms in a contract for drilling oil wells was explained in Startex Drilling Co. v. Sohio Petroleum Co., 680 F.2d 412 (5th Cir.1982):

It is more apt to say that the undefined technical terms on which the contract’s application to the present dispute depends convey little meaning without explanation. So, while we agree with Sohio that we are free to determine the ambiguity question anew, we also affirm the district court’s ruling that the contract is ambiguous. Thus it was proper to submit to the jury the evidence from both sides as to the meaning attached to these technical terms by the parties, and by the industry.

Id. at 415 (citation omitted). In Zell v. American Seating Co., 138 F.2d 641 (2d Cir.1943), Judge Frank wrote for the court that:

Thayer delightfully described the fatuous notion of a “lawyer’s Paradise, where all words have a fixed, precisely ascertained meaning; where men may express their purposes, not only with accuracy, but with fullness; and where, if the writer has been careful, a lawyer, having a document referred to him, may sit in his chair, inspect the text, and answer all questions without raising his eyes.”

Id. at 648 n. 26 (quoting Thayer, A Preliminary Treatise on Evidence (1898)).

It has not heretofore been seriously challenged that findings of the weight and credibility of evidence are for the jury, whether the issues are technologic, scientific, or otherwise. See Sartor v. Arkansas Natural Gas Corp., 321 U.S. 620 , 64 S.Ct. 724 , 88 L.Ed. 967 (1944):

[The weight and credibility of a witness’ testimony] “belongs to the jury, who are presumed to be fitted for it by their natural intelligence and their practical knowledge of men and the ways of men; and so long as we have jury trials they should not be disturbed in their possession of it, except in a case of manifest and extreme abuse of their function.”

Id. at 628 , 64 S.Ct. at 729 (quoting Aetna Life Ins. Co. v. Ward, 140 U.S. 76, 88 , 11 S.Ct. 720, 724 , 35 L.Ed. 371 (1891)). In Railroad Dynamics, Inc. v. A Stucki Co., 727 F.2d 1506 , 220 USPQ 929 (Fed.Cir.), cert. denied, 469 U.S. 871 , 105 S.Ct. 220 , 83 L.Ed.2d 150 (1984) this court instructed a party who sought de novo appellate review after a jury trial:

Thus [Railroad Dynamics] misconceives our role as an appellate court. In the concert hall of justice, each musician has a part to play. When one on whim plays not his own but another’s part, discord is certain. Moreover, our parts are played under well defined rules.

Id. at 1514, 220 USPQ at 937 .

Implementing this court’s departure from the established appellate role, in reviewing Markman’s case on this appeal the majority does not mention the jury instructions, or discuss whether a reasonable jury could have reached the verdict that was reached on the evidence adduced, or decide whether there was substantial credible evidence of such content and weight as could support the jury’s verdict. The majority finds for itself *1008 the disputed fact of whether the term “inventory” includes the invoices of the Westview system, without any deference to the trial process. Whether or not this court believes that it is a superior finder of technologic fact, that is not our place in the judicial structure.

I wonder how this new system will work. The majority states that the trial judge should have decided the meaning of “inventory” before giving the ease to the jury, 5 but that the error was harmless since the trial judge reached the correct meaning “as a matter of law” after the jury verdict. There was no return to the jury after the trial judge re-decided what “inventory” meant, making the jury superfluous. The Federal Circuit now decides de novo whether “inventory” includes “invoices,” ignoring the trial. What of the trial process, if trial judge and jury are ciphers upon appellate review? In the Introduction to the Reference Manual on Scientific Evidence (1994), Judge Schwarzer wrote:

The bedrock of [the justice] system is the adversary process, which depends on attorneys to present evidence on behalf of their clients, judges to make the necessary and appropriate rulings concerning admissibility, and juries to resolve disputed issues of fact.

Id. at 1.

In patent cases, no less than for other causes of action, it is the trier of fact on whom the system of justice is founded. The extensive exposition of disputed facts that is available at trial can not be duplicated on appeal. Even were there no constitutional infirmity, I can discern no practical benefit sufficient to justify this court’s departure from the established procedures of trial and appeal. Implicit in the appellate process is an expected degree of deference to the trial process. The majority’s elimination of the jury as trier of fact, and elimination of the deference owed to the judge upon bench trial of disputed facts, removes from the parties the benefit of the trial process. It distorts the trial/appellate relationship in a manner unique to patent litigation, and manifests a heady misperception of our assignment as a national appellate court.

C. THE “CLASSIFICATION POWER”— TURNING FACT INTO LAW

Commentators have remarked on the temptation of appellate courts to redefine questions of fact as questions of law in order to impose the court’s policy viewpoint on the decision. Professor Martin Louis calls the appellate assertion of power to treat fact as law “drastic in that it amounts to a direct judicial assault on the prerogatives of fact finders.” Martin B. Louis, Allocating Adjudicative Decision Making Authority Between the Trial and Appellate Levels: A Unified View of the Scope of Review, the Judge/Jury Question, and Procedural Discretion, 64 N.C.L.Rev. 993, 1018 (1986). Louis observes that the “classification of ultimate facts as questions of law amounts to a manipulation of the law-fact doctrine to take questions from the jury or to subject the trial level’s resolution of questions to free appellate review.” Id. at 1028. Although the Seventh Amendment has provided a safe *1009 guard against this autocracy of the judiciary, concerned observers have long counselled vigilance. Thus definitions of fact and law— the methodology of this appellate power— have attracted the attention of legal scholars.

“Law” is usually defined as a statement of the general principle or rule, predicated in advance, awaiting application to particular facts as they may arise. See Francis H. Bohlen, Mixed Questions of Law and Fact, 72 U.Pa.L.Rev. Ill, 112 (1924). Louis, supra, at 994, states the principle:

Declarations of law are fact-free general principles that are applicable to all, or at least to many, disputes and not simply to the one sub judice.

There is an additional element to “law;” that is, the duty of judicial enforcement. As Professor Thayer explained, “nothing is law that is not a rule or standard which it is the duty of judicial tribunals to apply and enforce.” James B. Thayer, “Law and Fact” in Jury Trials, 4 Harv.L.Rev. 147, 153 (1890).

Thayer defines “fact” as follows:

[“Fact”] is what Locke expresses when he speaks of “some particular existence, or, as it is usually termed, matter of fact.” The fundamental conception is that of a thing as existing, or being true. It is not limited to what is tangible, or visible, or in any way the object of sense; things invisible, mere thoughts, intentions, fancies of the mind, propositions, when conceived of as existing or being true, are conceived of as facts. The question of whether a thing be a fact or not, is the question of whether it is, whether it exists, whether it be true. All inquiries into the truth, the reality, the actuality of things are inquires into the fact about them.

Id. at 151-52. A compilation of definitions of “fact” is provided in Black’s Law Dictionary 591-92 (6th ed. 1990):

A thing done; an action performed or an incident transpiring; an event or circumstance; an actual occurrence; an actual happening in time or space or an event mental or physical; that which has taken place.

.... “Fact” means reality of events or things the actual occurrence or existence of which is to be determined by evidence.

In sum, the law is a general proposition, while the fact is a case-specific inquiry. Clarence Morris, Law and Fact, 55 Harv. L.Rev. 1303, 1304 (1942), observed that a controlling distinction between law and fact is whether evidence is needed, for a question of fact usually calls for proof, whereas matters of law are established not by evidentiary showing but by intellectual abstraction.

These distinctions have often been discussed, usually in the course of considering the complexities that can arise, and how they have been, or should be, treated. Thus commentators and judges have written to explain the distinctions among historical facts, ultimate facts, and mixed law/fact questions, in the course of relating these distinctions to trial procedures and judicial review. The nicety that has been generated was criticized in Armour & Co. v. Wilson & Co., 274 F.2d 143 , 124 USPQ 115 (7th Cir.1960), as follows:

We have come to speak of questions of “fact,” “primary facts,” “subsidiary facts,” “evidentiary facts,” “ultimate facts,” “physical facts,” “documentary facts,” “oral evidence,” “inferences,” “reasonable inferences,” “findings of fact,” “conclusions,” “conclusions of law,” “questions of fact,” “questions of law,” “mixed questions of law and fact,” “correct criteria of law,” and so on ad infinitum. The simple answer is that we are all too frequently dealing in semantics, and our choice of words does not always reflect the magic we would prefer to ascribe to them.

274 F.2d at 155 , 124 USPQ at 124-25 (footnote omitted). See generally Steven A. Chil-dress & Martha S. Davis, Federal Standards of Review (2d ed. 1992).

The character of what is a fact does not change, even in those special cases that have been held to warrant plenary appellate review. 6 The subject matter that the majority *1010 now designates as “law” — the disputed meaning and scope of technologic terms and words of art as used in particular inventions — is not law, but fact. On any definition of fact and law, the question of whether “inventory” as used in Markman’s Claim 1 means only clothing or can include invoices is a question of fact: on Thayer’s criterion of whether the fact exists; on Morris’ criterion of whether there is a need for evidence; on Bohlen’s inquiry of whether the meaning is specific to the situation sub judice. The meaning of “inventory” is specific to this invention, this patent, this claim, this system, this defendant. Its determination is for the trier of fact.

Ill

THE CONSTITUTION

The most egregious lapse in the majority’s ruling is its discard of the jury right in patent eases. As I said at the outset, patent infringement has been tried to a jury in the United States for two hundred years, and in England since at least 1623. Disputes concerning “letters patent” for inventions were tried in the English courts, as for other forms of letters patent, as I shall illustrate. Patent infringement trials at common law included determination of validity as well as infringement. Whatever version of “law/ fact” this court now chooses to adopt, it can not redact the history of jury trials. The judicial obligation to safeguard the constitutional right is not defeasible by calling a patent a “statute,” or otherwise diminishing the vitality of the Seventh Amendment.

Thus the court, sitting en banc to overrule its contrary precedent, removes the jury from its role as the trier of fact. That right is assured by the Seventh Amendment:

In Suits at common law, where the value in controversy shall exceed twenty dollars, the right of trial by jury shall be preserved, and no fact tried by a jury shall be otherwise re-examined in any Court of the United States, than according to the rules of the common law.

U.S. Const, amend. VII.

The importance of the jury right to the Framers can not be overemphasized. Alexander Hamilton wrote:

The friends and adversaries of the plan of the convention, if they agree in nothing else, concur at least in the value they set upon the trial by jury; or if there is any difference between them it consists in this: the former regard it as a valuable safeguard to liberty; the latter represent it as the very palladium of free government.

The Federalist No. 83, at 499 (Clinton Rossi-ter ed., 1961). In discussing the history of the jury in England and in the United States, Judge Arnold has explained:

It is almost impossible to exaggerate the centrality of the institution of the jury to almost all the important episodes of Anglo-American legal history. Many of the central ideas of the American and English common law owe their origin to the fact that the jury was the chief mechanism for trying factual disputes. It is the single most important institution in the history of Anglo-American law.

Morris Sheppard Arnold, The Civil Jury in Historical Perspective, in The American Civil Jury 9, 10 (1987).

The value that the Framers placed on this “palladium of free government” has been guarded by the courts:

Maintenance of the jury as a fact-finding body is of such importance and occupies so firm a place in our history and jurisprudence that any seeming curtailment of the right to a jury trial should be scrutinized with the utmost care.

Dimick v. Schiedt, 293 U.S. 474, 486 , 55 S.Ct. 296, 301 , 79 L.Ed. 603 (1935).

The deference that courts give to jury verdicts is the mechanism by which the Constitution protects the jury right from encroachment by judges. It is not this court’s *1011 option to violate that right, whether by denying such deference or by taking from the jury the trial of factual issues. Whatever one’s personal view of the relative capabilities of a jury and the Federal Circuit in finding technologic facts in patent cases, it is not within our authority to readjust that role to our taste:

The Seventh Amendment ... requires that questions of fact in common law actions shall be settled by a jury, and that the court shall not assume directly or indirectly to take from the jury or to itself such prerogative.

Walker v. New Mexico & So. Pac. R. Co., 165 U.S. 593, 596 , 17 S.Ct. 421, 422 , 41 L.Ed. 837 (1897).

The Federal Circuit early affirmed its inheritance of this responsibility, as we undertook our assignment to provide nationwide uniformity in patent cases. In Connell v. Sears, Roebuck & Co., 722 F.2d 1542, 1547 , 220 USPQ 193, 197 (Fed.Cir.1983) the court stated:

So long as the Seventh Amendment stands, the right to a jury trial should not be rationed, nor should particular issues in particular types of cases be treated differently from similar issues in other types of eases.

In Railroad Dynamics v. Stucki, 727 F.2d at 1515 , 220 USPQ at 937 , the court stated:

There is, of course, no reason for considering patent cases as somehow out of the mainstream of the law and rules of procedure applicable to jury trials for centuries under our jurisprudence.

In many ensuing decisions we reaffirmed this obligation. However, this court’s fidelity to fundamental law slipped in recent years, culminating in today’s trivializing of our heritage as we defeat the jury right in patent infringement cases.

A. THE HISTORICAL TEST

There are no fine lines to be drawn in interpreting the Seventh Amendment, for all cases at common law in England were tried to a jury. In explaining the “historical test,” Justiee Story described England as “the grand reservoir of all our jurisprudence”:

Beyond all question, the common law here alluded to [in the Seventh Amendment] is not the common law of any individual state, (for it probably differs in all), but is the common law of England, the grand reservoir of all our jurisprudence. It cannot be necessary for me to expound the grounds for this opinion, because they must be obvious to every person acquainted with the history of the law.

United States v. Wonson, 28 F.Cas. 745, 750 (1812). The historical test assured the largeness of the embrace of the Amendment. I can find no support in history for the restriction today adopted. In England in 1791, as of at least 1623, actions of the “force and validity” of letters patent were tried according to the rules of the common law.

Letters patent were grants of the Crown, made for a variety of purposes. During the 1500s to early 1600s, the Star Chamber considered all infringements of letters patent to be contempts of royal authority. See Millar v. Taylor, 4 Burr. 2303, 2374 (K.B.1769) (contempt proceeding applied by the Star Chamber to infringement of “any patent the Crown thought proper to grant”); Coke, 3 Inst. 182-83 (discussing abuses). In 1623 the Statute of Monopolies prohibited all monopolies except patents for inventions, which continued to be granted for terms that were limited to fourteen years. Section 6 of the Statute stated:

6. Provided also, that any declaration before mentioned, shall not extend to any letters patents and grants of privilege for the term of fourteen years or under, hereafter to be made, of the sole working or making of any manner of new manufactures within this realm to the true and first inventor and inventors of such manufactures, which others at the time of making such letters patents and grants shall not use, so as also they be not contrary to the law nor mischievous to the state by raising prices of commodities at home, or hurt of trade, or generally inconvenient ...

21 Jac. I, e. 3, s. 6 (1623). 7 Section 2 of the Statute provided that the “force and validity” *1012 of the subject matter of the Statute shall be “examined, heard, tried, and determined” according to the common law:

2. And all monopolies, and all such commissions, grants, licenses, charters, letters patent, proclamations, inhibitions, restraints, warrants of assistance, and all other matters and things tending as aforesaid, and the force and validity of them, and every of them, ought to be, and shall be for ever hereafter examined, heard, tried, and determined, by and according to the common laws of this realm, and not otherwise.

21 Jac. I, c. 3, s. 2 (1623). Lord Coke explained that the purpose of Section 2 was to remedy the “mischief’ of Star Chamber actions by placing the authorized grants under the common law. 3 Inst, at 188.

The litigation procedures that applied to letters patent for inventions did not differ from those applied to other letters patent. See Benjamin Vaughan Abbott, Decisions on the Law of Patents for Inventions Rendered by English Courts 8 n. 2 (1887) (the rules for letters patent stated in the case of Rex v. Mussary (K.B.1738) were also generally applicable to letters patent for inventions). English cases of the period show similar procedures whether the subject was a charter, an invention, a literary work, a trademark, an interest in land, or a trade route.

Trial by jury was the way of the common law, and did not depend on the subject of the letters patent. See, e.g., East India Company versus Sandys, 1 Vern. 127 (Ch.1682) (validity of grant of exclusive trade route tried to jury); Mayor of Kingston upon Hull versus Horner, 1 Cowp. 102, 108 (K.B.1774) (dispute concerning meaning of terms of a charter from the Crown relating to a port “most proper to be left to the decision of the jury”); Blanchard v. Hill, 2 Atk. [2d ed. 1794] 484 (Ch.1742) (injunction to restrain the use of a tradesman’s mark granted by the Crown could not be issued without a hearing at law); Anon, 1 Vern. 120 (Ch.1682) (the Lord Keeper (or Chancellor) required trial at law to determine the validity of letters patent for printing of the Bible); Collins v. Sawrey, 4 Bro.P.C. [2d ed. 1803] 692 (H.L.1772) (rejecting argument that since the issue depended on written evidence, the “construction” of the letters patent for a vicarage was for the chancery court); Donaldson v. Beckett, 2 Bro. P.C. [2d ed. 1803] 129, 138 (H.L.1774) (“Between such inventions and copies of books, no sensible distinction can be drawn.”); Millar v. Taylor, 4 Burr, at 2323:

In letters patent, all conditions required by 21 Jac. 1 must be observed. Patentees for new inventions are left, by that statute, to the common law, and the remedies which follow in their nature.

There simply is no way, in 1995, to rewrite the history of England and the place of the jury under the common law in 1791.

Actions for infringement of letters patent for inventions were initiated either in the law courts or in Chancery, depending on the relief sought. 8 When equitable relief was sought, patent actions began with the filing of a bill in the Ordinary side of Chancery (called the Petty Bag or Latin Side), for that *1013 is where the 'letters patent were “enrolled.” Issues relating to invalidity and noninfringement, if raised by the defendant, were directed by the Chancellor to the courts of law for trial to a jury, and then returned to Chancery if the verdict warranted. The procedure of filing a bill in Chancery and then trying the issue at law was explained by Davies:

[T]he Court of Chancery never decides upon the validity of a patent, the practice there being nothing more than to grant an injunction, at the prayer of the patentee, against any person infringing his patent, and to order an account of profits; but if any question arises upon the validity of the patent, novelty of the invention, or the sufficiency of the specification, it is uniformly referred to a court of law.

John Davies, A Collection of the Most Important Cases Respecting Patents of Invention and the Rights of Patentees ix (1816).

In infringement suits, the Chancery court could grant the patentee’s bill seeking an injunction, or a writ of scire facias to repeal the patent, after trial to a jury in a court of law. See, e.g., Brewster v. Weld, 6 Mod. 229 (1704) (a scire facias to repeal letters patent may be sued in Chancery by any person prejudiced by a patent, as well as by the Crown; when Chancery issues writ returnable to Queen’s Bench [requiring trial to a jury] Chancery neither has jurisdiction nor can it supersede such writ); Rex v. Else, 1 Carp.P.C. 103 (K.B., N.P. 1785) (proceeding brought by writ of scire facias to repeal patent on ground that there was no new invention described in the specification; tried in King’s Bench wherein the jury rendered a verdict for the Crown). Not all matters required the Chancellor to direct issues to the law courts to be tried. For example, when the Crown granted letters patent, for invention or otherwise, the grant had to be enrolled in Chancery’s Petty Bag Office within four months for the patent to be enforceable. Thus a bill could be filed in Chancery to seek equitable relief for a patentee’s failure to enroll the patent before the time expired. E.g., Ex parte Beck, 1 Bro.P.C. [2d ed. 1803] 578 (Ch.1784).

These relationships were well established by the date of the Seventh Amendment. Issues of patent infringement and validity were tried only to a jury, in the courts of King’s Bench, Common Pleas, or Assize. In a common procedure the patentee would seek an injunction against infringement, the defendant would assert invalidity, and the matter would be directed to a court of law for trial. This process is illustrated in Newsham v. Gray, a patent infringement action that started with a bill in equity, seeking to enjoin Gray, the alleged infringer. The Lord Keeper directed the plaintiff to bring an action at law. The following is from Lord Chancellor Hardwicke’s opinion in the subsequent proceeding in Chancery, where Gray was seeking to recover costs since the plaintiff was nonsuited for failure to prosecute:

The plaintiff had obtained letters patent of the crown for a new invention of fire engines.

A bill was brought by him to establish his letters patent, and for a perpetual injunction against the defendant, who had taken upon him to make and vend these engines, notwithstanding the plaintiff had sole right and property under the letters patent.

The defendant, by his answer, insisted it was not a new invention, so as to entitle the plaintiff to an injunction.

There was no replication, but the cause came on at the Rolls, upon bill and answer, in September 1740, before Mr. Justice Parker, who, not thinking the answer sufficient, directed an action at law to be brought by the plaintiff, for a breach of the letters patent, and retained the bill for a twelvemonth; the plaintiff was nonsuited at law upon the merits; and the cause is now set down by the defendant for a dis-mission of the bill, and for costs.

2 Atk. [2d ed. 1794] 286, 286-87 (Ch.1742).

When the patentee did not seek equitable relief, the action was brought directly at law. The cause of action was trespass on the case. The action was an offspring of the criminal law, and knew no form but trial by jury. See H.G. Hanbury and D.C.M. Yardley, English Courts of Law 64 (5th ed. 1979) (1944). The defendant could assert defenses including in *1014 validity and noninfringement. All issues, including damages, were for the jury.

The burden of proof was on the patentee. Since letters patent of invention were issued without examination, simply upon declaration, actions to enforce the patent began with proof of entitlement to the patent, if disputed by the defendant. The burden of proving infringement was also on the plaintiff, if infringement was disputed. The English reports show that often patent infringement actions turned on the issue of entitlement or validity, whereupon when validity was found, verdict would be rendered for infringement. For example, Dolland’s Case, 1 Carp.P.C. 28 (C.P.1766) 9 was an action for trespass on the case, seeking damages resulting from infringement of Dolland’s 1758 patent on a telescope. The defendant asserted invalidity due to prior use. The jury verdict was for Dolland.

Morris v. Bramson, 1 Carp.P.C. 30 (K.B.1776) was an action for infringement of Morris’ patent, a patent previously tried and adjudged valid and infringed in Morris v. Else. (Morris v. Else is unreported in the English Reports; the ease is discussed in Boulton v. Bull, 2 H.B1. at 489.) In Morris v. Bramson the defendant argued that an addition to an old machine was not patentable as a matter of law. The judge instructed the jury on the law, and the jury found for the plaintiff, awarding 500 pounds in damages for infringement. 1 Carp.P.C. at 31.

Bramah v. Hardcastle, 1 Carp.P.C. 168 (K.B.1789), was an action in trespass on the case for infringement of letters patent for a new construct of a water closet. The defendant asserted invalidity due to prior use and lack of novelty. Lord Kenyon is reported as telling the jury “the patent was void, the invention not being new,” id. at 171, and that they should find for the defendant. 10 The jury sustained the patent, and found infringement. The court entered judgment in accordance with the jury verdict.

Arkwright v. Nightingale, 1 Carp.P.C. 38 (C.P.1785), was an action for infringement of a 1775 letters patent for “machines of utility in preparing silk, cotton, flax and wool for spinning.” At trial the defendant claimed that the patent was invalid because of an inadequate disclosure in the specification. At the close of the trial, Lord Loughborough provided a lengthy summary of the evidence, and concluded his charge with: “Therefore the single question is, whether you believe these five witnesses are peijured, or that they speak the truth. According as you are of the opinion, one way or the other, you will find your verdict for the plaintiff or the defendant.” 1 Carp.P.C. at 53. The jury rendered a verdict for the plaintiff, that is, infringement by the defendant.

Rex v. Arkwright, 1 Carp.P.C. 53 (K.B.1785): After the decision in Arkwright v. Nightingale, supra, a scire facias was filed with the High Court of Chancery to repeal the patent from the rolls, the petitioner asserting in part that the invention was not new as to use in England, and that Arkwright was not the inventor. The issue was tried to a jury in King’s Bench. After the close of the evidence, the court instructed the jury:

Gentlemen, thus the case stands as to the several component parts of this machine; and if upon them you are satisfied none of them were inventions unknown at the time this patent was granted, or that they were not invented by the defendant; upon either of these points the prosecutor is entitled to your verdict.

1 Carp.P.C. at 101. The jury found for the prosecutor.

Turner v. Winter, 1 T.R. 602 (K.B.1787), reports a ruling on a motion to set aside a jury verdict of patent infringement and grant *1015 a new trial. The court granted the motion, explaining that:

And if it appear that there is any unnecessary ambiguity affectedly introduced into the specification, or anything which tends to mislead the public, in that case the patent is void. Here it does appear to me, that there is at least such a doubt on the evidence, that I cannot say this matter has been so fully and'fairly examined, as to preclude any further investigation of the subject.

1 T.R. at 605. The case was remanded for a new trial.

Administrators of Calthorp v. Waymans, 3 Keb. 710 (K.B.1676) was an action for infringement of a patent on an engine. The jury was instructed that English law required novelty only in England, and did not require that an importer/patentee of a device new to England be the actual inventor. The jury found for the patentee, the report of the case explaining that “it appeared in evidence to a jury at Bar, that the fashion came out of Holland,, and was there used above fifty years since, but never before used in England.” 3 Keb. at 710.

I again stress that actions at law were tried to a jury. With respect to letters patent for inventions, and in accordance with the Statute of Monopolies, in seventeenth- and eighteenth-century England patent infringement was tried to a jury at common law. I have come upon no exception in the cases reported during this period.

B. CONTINUITY IN THE UNITED STATES

The reports of English patent cases do not manifest the turmoil in preserving the jury right in England, the imprisonment of jurors before 1670, and attempts to limit the jury right in England as well as in the American colonies. Reflecting this experience, there was in the new United States a reverence for the place of the jury as, in the words of Thomas Jefferson, “the only anchor yet imagined by man, by which a government can be held to the principles of its constitution.” Parklane Hosiery Co. v. Shore, 439 U.S. 322 , 343 n. 10, 99 S.Ct. 645 , 658 n. 10, 58 L.Ed.2d 552 (1979) (Rehnquist, J., dissenting) (quoting 3 The Writings of Thomas Jefferson 71 (Washington ed. 1861)).

The Supreme Court summarized a long history in the statement:

The trial by jury is justly dear to the American people. It has always been an object of deep interest and solicitude, and every encroachment upon it has been watched with great jealousy.

Parsons v. Bedford, 28 U.S. (3 Pet.) 433, 445 , 7 L.Ed. 732 (1830). Justice Story made clear that the right was not limited to the precise causes of action that existed in the law courts of England. Id., 28 U.S. at 446-47 . See generally James Fleming, Jr., Right to a Jury Trial in Civil Actions, 72 Yale L. J. 655 (1963). In Curtis v. Loether, 415 U.S. 189 , 94 S.Ct. 1005 , 39 L.Ed.2d 260 (1974) the Court wrote:

Although the thrust of the Amendment was to preserve the right to jury trial as it existed in 1791, it has long been settled that the right extends beyond the common-law forms of action recognized at that time.

415 U.S. at 193 , 94 S.Ct. at 1007 . In Tull v. United States, 481 U.S. 412, 417 , 107 S.Ct. 1831, 1835 , 95 L.Ed.2d 365 (1987) the Court reiterated that the right to jury trial extends to causes of action created by Congress which are similar to common law forms of action. As a recent example, in Chauffeurs, Teamsters, and Helpers Local No. 391 v. Terry, 494 U.S. 558 , 110 S.Ct. 1339 , 108 L.Ed.2d 519 (1990), the Court considered whether a suit in which an employee sought back pay, for breach of a union’s duty of fair representation, carried the right to a jury trial. The Court stated:

To determine whether a particular action will resolve legal rights, we examine both the nature of the issues involved and the remedy sought. “First, we compare the statutory action to 18th-century actions brought in the courts of England prior to the merger of the courts of law and equity. Second, we examine the remedy sought and determine whether it is legal or equitable in nature.” Tull, supra [481 U.S.], at 417-18 [ 107 S.Ct. at 1835 ]'(citations omitted). The second inquiry is the more im *1016 portant in our analysis. Granfinanciera, S.A. v. Nordberg, 492 U.S. 33, 42 [ 109 S.Ct. 2782, 2790 , 106 L.Ed.2d 26 ] (1989).

494 U.S. at 565 , 110 S.Ct. at 1345 (footnote omitted). Observing that the cause of action of a union’s duty was unknown in eighteenth-century England, the Court looked to analogous actions, including an action to set aside an arbitration award, an action of a beneficiary against a trustee, and an attorney malpractice action. Id. at 565-66 , 110 S.Ct. at 1344-45 . The Court held that the respondents were entitled to a jury trial under the Seventh Amendment, despite the equitable nature of the underlying action, since the relief sought was legal in nature. See, e.g., Beacon Theatres, Inc. v. Westover, 359 U.S. 500 , 79 S.Ct. 948 , 3 L.Ed.2d 988 (1959):

As this Court said in Scott v. Neely, 140 U.S. 106, 109-110 [ 11 S.Ct. 712, 714 , 35 L.Ed. 358 (1891) ]: “In the Federal courts this [jury] right cannot be dispensed with, except by the assent of the parties entitled to it, nor can it be impaired by any blending with a claim, properly cognizable at law, of a demand for equitable relief in aid of the legal action or during its pendency.”

359 U.S. at 510 , 79 S.Ct. at 956-57 (footnote omitted). See also Atlas Roofing Co. v. Occupational Safety and Health Review Comm’n, 430 U.S. 442 , 97 S.Ct. 1261 , 51 L.Ed.2d 464 (1977).

On this history, it is jarring to come upon the majority’s argument that the Seventh Amendment no longer applies because there are now “claims” in United States patents, whereas the old English patents did not have claims as we know them. The removal of the jury right is not so casually achieved:

[T]he Constitution is concerned, not with form, but with substance. All of vital significance in trial by jury is that issues of fact be submitted for determination with such instructions and guidance by the court as will afford opportunity for that consideration by the jury which was secured by the rules governing trials at common law.

Gasoline Prods. Co. v. Champlin Ref. Co., 283 U.S. 494, 498 , 51 S.Ct. 513, 514 , 75 L.Ed. 1188 (1931).

However, the argument about claims does bring out a point of curiosity, for the law of eighteenth-century England required specificity in “particularly describing” what was patented, and the patent grant ended with a concise summary of the subject matter, with details annexed in the specification; patent “claims,” in turn, are concise summaries of the subject matter, with details annexed in the specification. Following is a portion of a representative letters patent dated March 28, 1764:

To all to whom these presents shall come, John Morris, of the town of Nottingham, hosier, sendeth greeting.—

Whereas, the King’s Most Excellent Majesty, by letters patent under the Great Seal of Great Britain, bearing the date at Westminster, [gave and granted to the inventors] sole privilege ... to make, use, exercise, and vend their invention ... in which said letters patent is contained a proviso that if the said Thomas and John Morris, and John and William Betts, or any one of them should not particularly describe the nature of the said invention and in what manner the same is to be performed by an instrument, in writing under their hands and seals, or the hand and seal of one of them, and cause the same to be enrolled in the High Court of Chancery ...

[The specification ending with:] Now know ye, that I, the said John Morris, in pursuance of the said proviso in the said letters patent contained, do hereby declare that the said invention of an engine or machine, on which is fixed a set of working needles, which engine or machine is fixed to a stocking-frame for the making of oilet-holes or net-work in silk, thread, cotton, or worsted, as mitts, gloves, hoods, aprons, handkerchiefs, and other goods usually manufactured upon stocking-frames by a method entirely new, is particularly described in the plans hereunderto annexed.

Morris v. Brojmson, 1 Carp. P.C. at 31-32 n. * (emphasis added). The requirement that the inventor “particularly describe” the invention was carried into the United States Patent Act of 1790:

*1017 Sec. 2. And be it further enacted, That the grantee or grantees of each patent shall, at the time of granting the same, deliver to the Secretary of State a specification in writing, containing a description, accompanied with drafts or models, and explanations and models (if the nature of the invention or discovery will admit of a model) of the thing or things, by him or them invented or discovered, and described as aforesaid, in the said patents; which specification shall be so particular, and said models so exact, as not only to distinguish the invention or discovery from other things before known and used, but also to enable a workman or other person skilled in the art or manufacture, whereof it is a branch, or wherewith it may be nearest connected, to make, construct, or use the same....

Patent Act of 1790, ch. 7, § 2, 1 Stat. 109,110 (1790) (emphasis added).

This requirement was continued in all subsequent revisions, which were successively more explicit. In 1836 the Patent Act required that the inventor “particularly specify the part, improvement, or combination which he claims as his own invention.” Ch. 357, § 6, 5 Stat. 117 , 119. Again revised in 1870, the statute required that the inventor “particularly point out and distinctly claim the part, improvement, or combination which he claims as his invention or discovery.” Ch. 230, § 26, 16 Stat. 198 , 201. The present statute, enacted in 1952, states that “the specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.” 35 U.S.C. § 112 , ¶2. This evolution in statutory directive, requiring the inventor to be more specific as to what had been invented, did not remove the jury from trial of patent infringement cases.

The majority’s other response to the Constitution is to call a patent a “statute,” arguing that “statutory interpretation” is not for the jury. Designating a patent a statute in order to avoid the Seventh Amendment simply denies history and our heritage. Our judicial responsibility is to uphold the Constitution, not devise ways to circumvent it.

IV

PRECEDENT

A. FEDERAL CIRCUIT CASES SELECTED FOR CRITICISM AND OVERRULE

The Federal Circuit early in its existence deplored the “risk of effectively denying the constitutional right spelled out in the first clause of the Seventh Amendment.” Railroad Dynamics v. Stucki, 727 F.2d at 1515 , 220 USPQ at 937-38 . Many Federal Circuit decisions implemented the correct standard of trial and appellate review in patent infringement cases. The majority now expressly disapproves appellate deference to the trier of fact on the issues of fact that are determined in the course of “construing” the meaning and scope of patent claims, issues of fact that are dispositive of the question of patent infringement. The majority singles out seven cases for specific criticism, and fatally taints the many other cases that applied the correct standard of deference to the trier of fact.

The majority explains that it overrules these cases because this court held that the interpretation of disputed technologic terms in patent claims raises jury-triable issues, or because the panel applied a deferential standard of appellate review. Majority op. at 993-994. The majority does not tell us how such cases will be tried, now that appeal includes mandatory de novo adjudication of what were once recognized as triable facts. Even the least cynical observer must wonder at the court’s capacity for this technological overload. A glance at the subject matter of the seven expressly disapproved cases illustrates these problems.

1. McGill, Inc. v. John Zink Co., 736 F.2d 666 , 221 USPQ 944 (Fed.Cir.), cert. denied, 469 U.S. 1037 , 105 S.Ct. 514 , 83 L.Ed.2d 404 (1984).

The disputed technical term in the patent claim was “recovered liquid hydrocarbon absorbent.” On McGill’s view of what this term meant, the Zink process would infringe McGill’s claim; on Zink’s view, the Zink pro *1018 cess would not infringe. There was conflicting testimony of technical experts, and the issue was submitted to the jury. On appeal the Federal Circuit made the now-excoriated statement:

If, however, the meaning of a term of art in the claims is disputed and extrinsic evidence is needed to explain the meaning, construction of the claims could be left to the jury. In the latter instance, the jury cannot be directed to the disputed meaning for the term of art.

Id. at 672, 221 USPQ at 948 (citations omitted). On appellate review the court considered the meaning of the claims upon the following criterion:

In the instant case, the jury’s finding of infringement was predicated on construction of claim 2. To obtain a reversal, Zink must demonstrate that no reasonable juror could have interpreted the claim in the fashion that supports the infringement finding.... Zink must convince us that there is no set of facts, consistent with McGill’s interpretation, that was supported by substantial evidence.

Id. The majority now holds that the meaning of “recovered liquid hydrocarbon absorbent” and the other disputed technical terms that were at issue was not a jury triable issue, and that the Federal Circuit should have, and hereafter will, decide such questions as a matter of law.

2. Bio-Rad Labs., Inc. v. Nicolet Instrument Corp., 739 F.2d 604 , 222 USPQ 654 (Fed.Cir.), cert. denied, 469 U.S. 1038 , 105 S.Ct. 516 , 83 L.Ed.2d 405 (1984).

The patented device was an interferometer that contained an oscillating mirror that varied the lengths of two of four possible paths of split beams of reflected light, whereby the thickness of the epitaxial layer of a semiconductor was determined from the points of locally maximum constructive interference, by comparing phase differences. The jury trial lasted forty-four days. On appeal this court stated that we review to determine

whether reasonable jurors, after reviewing all the evidence, could have interpreted the claims to include the sequence of events followed by [the accused optical apparatus].

Id. at 613, 222 USPQ at 661 (citing Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 893 , 221 USPQ 669, 673 (Fed.Cir.), cert. denied, 469 U.S. 857 , 105 S.Ct. 187 , 83 L.Ed.2d 120 (1984)). This court declined the losing party’s request that we make a de novo interpretation of the claims:

We emphasize that our task is not to interpret the claims as though no trial occurred. Both parties submitted testimony in support of their interpretation before the jury. Bio-Rad’s interpretation prevailed and was not overturned by the trial judge. On appeal, we consider only whether reasonable jurors could have interpreted the claim in the manner presumed.

Id. at 614, 222 USPQ at 661-62 . This practice can no longer be followed, and the Federal Circuit shall somehow conduct these technological analyses for ourselves, as a matter of law.

3. Palumbo v. Don-Joy Co., 762 F.2d 969 , 226 USPQ 5 (Fed.Cir.1985).

This case too is criticized for its holding that the findings of disputed facts of the meaning of claim terms is for the trier of fact. The appeal reached us on summary judgment. Palumbo sued Don-Joy for infringement of a patent to a patellar brace used in diagnosis and treatment of patellar subluxation (dislocation of the kneecap). In holding that summary judgment was improperly granted, this court referred to the disputed factual issues that had been raised in the depositions, as well as ambiguity in the prosecution history and the need for expert witnesses to present the viewpoint of those of skill in this art. The court stated:

If the language of a claim is not disputed, then the scope of the claim may be construed as a matter of law. But when the meaning of a term in a claim is disputed and extrinsic evidence is necessary to explain that term, then an underlying factual question arises, and construction of the claim should be left to the trier or jury under appropriate instruction.

Id. at 974 , 226 USPQ at 8 . Overruling the statement that the meaning of claim terms *1019 can raise underlying factual questions and that disputed “claim construction” should be left to the trier or jury, the majority now requires that the Federal Circuit shall make these decisions de novo.

4. Moeller v. Ionetics, Inc., 794 F.2d 653 , 229 USPQ 992 (Fed.Cir.1986).

As I mentioned supra, the majority also disapproves this case, criticizing its holding that “disputes over the meaning of claim language may raise factual questions reviewed for substantial evidence or clear error as the case may be,” in the majority’s words. The invention was a system of selectively measuring the concentration of certain cations in the presence of other components, by interposing a membrane barrier and using specified electrodes whereby cation-specific components such as nonaetin, gramicidin, and valinomycin form positively charged complexes with the sensing device. The disputed term claims were “electrode,” “electrode body,” and “disposed in said body.” The technologic meaning of these terms, in this usage and this invention, decided whether the terms encompassed the accused system. On appeal this court observed that the meanings of these terms were “clearly disputed,” referring to conflicting evidence, and held that the matter required trial, vacating the grant of summary judgment. It appears that this court will now decide what these terms mean as a matter of law.

5. H.H. Robertson Co. v. United Steel Deck, Inc., 820 F.2d 384 , 2 USPQ2d 1926 (Fed.Cir.1987).

In connection with a motion for preliminary injunction, the dispositive issue before the district court was the meaning of “bottomless trench” in a patent for a concrete deck structure for distributing electrical wiring. The defendants argued that their structure was not “truly bottomless” because “horizontal metal sections” and a “horizontal metal strip” constituted a partial bottom. After a four-day hearing, the district court granted the preliminary injunction. In affirming, this court described its standard of review:

Claim construction is reviewed as a matter of law. However, interpretation of a claim may depend on evidentiary material about which there is a factual dispute, requiring resolution of factual issues as a basis for interpretation of the claim. In this case, there was extensive testimony on the issue of claim construction, including the conflicting views of experts on both legal and factual questions. Those factual considerations that are pertinent to the district court’s construction of the term “bottomless” are reviewed under the clearly erroneous standard.

Id. at 389 , 2 USPQ2d at 1929 (citations omitted). The majority condemns this case for its recognition that “claim construction” may require resolution of factual issues, and the use of the clearly erroneous standard of review for those factual findings. Not only juries, but trial judges, will now be denied the deference owed to their factual findings.

6. Perini America, Inc. v. Paper Converting Mach. Co., 832 F.2d 581 , 4 USPQ2d 1621 (Fed.Cir.1987).

This too was a bench trial, and had been fully tried to the court. As I mentioned supra, the dispute related to various aspects of machines used in manufacturing paper towels. This court recognized that these were factual issues, reviewed on the clearly erroneous standard:

A trial court’s conclusions on the scope of the claims are reviewable as matters of law, but findings on disputed meanings of terms in the claims and on the infringement issue must be shown to have been clearly erroneous.

Id. at 584, 4 USPQ2d at 1624 (citations omitted). The court observed that “legal conclusions [are] dictated by established facts and not the other way around, and does not change the nature of the meaning-of-terms inquiry from one of fact to one of law.” Id. The majority strongly criticizes, and overrules, these statements.

7. Tol-O-Matic, Inc. v. Proma Produkt-Und Marketing Gesellschaft m.b.H, 945 F.2d 1546 , 20 USPQ2d 1332 (Fed.Cir.1991).

This was a jury trial. The technology related to rodless piston-cylinders. The in *1020 vention was for a yoke structure that reduces the forces tending to widen the slit through which the external load is moved by the piston, thereby avoiding loss of cylinder pressure. At issue was the meaning of the term “to provide for lateral support of the portions of the cylinder separated by the slit and spanned by the yoke.” The decision required a choice between Tol-O-Matic’s position that this term meant that the yoke must prevent all widening of the slit, and Proma’s position that this term required only some resistance to slit widening. At the trial there was testimony by engineers representing both sides, who ran tests on rodless cylinders under various conditions and reached inconsistent results. The jury was instructed to consider all of the evidence and find the meaning of the disputed term, and then to apply it to the accused device. On appeal this court endorsed the procedure:

The interpretation of claims is defined as a matter of law based on underlying facts. Interpretation of the claim words “provide for lateral support” required that the jury give consideration and weight to several underlying factual questions, including in this case the description of the claimed element in the specification, the intended meaning and usage of the claim terms by the patentee, what transpired during the prosecution of the patent application, and the technological evidence offered by the expert witnesses. When the meaning of a term in a patent claim is unclear, subject to varying interpretations, or ambiguous, the jury may interpret the term en route to deciding the issue of infringement. The jury’s verdict of noninfringement is reviewed, in accordance with the rules governing review of jury determinations, to ascertain whether reasonable jurors could have interpreted the claim in a way that supports the verdict.

Id. at 1549-50 , 20 USPQ2d at 1335-36 (emphases added) (citations omitted).

This case is severely criticized for the emphasized statements. According to the majority these questions are not factual, can not be given to the jury, are not reviewed with deference to the trier of fact, and will be decided de novo by the Federal Circuit.

B. OTHER IMPUGNED FEDERAL CIRCUIT CASES

The court in Tol-O-Matic cited Tillotson, Ltd. v. Walbro Corp., 831 F.2d 1033 , 4 USPQ2d 1450 (Fed.Cir.1987); Tandon Corp. v. United States Int’l Trade Comm’n, 831 F.2d 1017 , 4 USPQ2d 1283 (Fed.Cir.1987); Howes v. Medical Components, Inc.; Moeller v. Ionetics, Inc.; Snellman v. Ricoh Co., Ltd., 862 F.2d 283 , 8 USPQ2d 1996 (Fed.Cir.1988), ce rt. denied, 491 U.S. 910 , 109 S.Ct. 3199 , 105 L.Ed.2d 707 (1989); Vieau v. Japax, Inc., 823 F.2d 1510 , 3 USPQ2d 1094 (Fed.Cir.1987); Data Line Corp. v. Micro Technologies, Inc., 813 F.2d 1196 , 1 USPQ2d 2052 (Fed.Cir.1987); Palumbo v. Don-Joy Co.; Bio-Rad Labs., Inc. v. Nicolet Instrument Corp.; Perkin-Elmer Corp. v. Computervision Corp.; and McGill, Inc. v. John Zink Co.. Only some of these decisions are today singled out for criticism; but all explicitly recognized the now-rejected difference between fact and law as applied to the meaning of disputed terms in patent claims, and all deferred, on appellate review, to the findings of the trier of fact.

For example, Data Line v. Micro Technologies related to computer technology wherein the jury, hearing expert testimony, interpreted “means for sensing the presence or absence of output data”; this court on appeal rejected the appellant’s argument that the trial court should have “determine[d] the scope and construction of claim 1,” and instead gave deferential review to the jury verdict. In Snellman v. Ricoh the jury verdict was reviewed on the substantial evidence standard, not de novo. In Delta-X Corp. v. Baker Hughes Production Tools, Inc., 984 F.2d 410, 415 , 25 USPQ2d 1447, 1450 (Fed.Cir.1993), this court approved the trial procedure whereby “because of disputes over claim terms, the judge instead left resolution of these disputes to the jury.”

There are many more cases than those I have listed, in which the jury decided technological and other factual disputes concerning the meaning and scope of terms of patent claims, thereby also deciding the fact of infringement, and where the jury verdict was reviewed on the usual substantial evi *1021 dence/reasonable jury standard. There are many more eases than those I have listed, in which the district court at bench trial found the facts of what the claim terms mean and cover, and on appellate review this court applied the clearly erroneous standard of review. These procedures, which are in accord with factual determinations in other areas of litigation, have now been rejected. The new and unique treatment of disputed facts in patent cases does not appear to offer advantages to outweigh its disadvantages. 11

C. THE MAJORITY’S CITED AUTHORITY

The authority on which the majority relies simply does not support its statement that “the Supreme Court has repeatedly held that construction of a patent claim is a matter of law exclusively for the court.” Majority op. at 994. That statement is of course correct when deciding the legal effect of a patent claim, and when stating the law to be applied by the trier of fact in interpreting disputed terms. However, it is not correct with respect to findings of disputed factual issues, issues that usually relate to the meaning and scope of the technologic terms and words of technical art that define the invention. Even the majority’s selected authority recognized that such issues are factual, to be found by the jury. Although the majority now equates these factual findings with “construction of a patent,” the Supreme Court did not.

In Winans v. Demnead, 56 U.S. (15 How.) 330 , 14 L.Ed. 717 (1854) the invention was the conical shape of a coal-carrying railroad car, whereby the car could carry several times its weight in coal. The car described and claimed in Winans’ patent had a circular cross-section; the accused ear of Denmead had an octagonal cross-section. The trial court instructed the jury that since the patent described the circular shape, it was so limited. The Supreme Court held that the instruction was in error, and that the jury should have been instructed on the legal rule that the thing patented was not limited to the exact shape or form illustrated, but depended on whether the same function was performed in substantially the same way and with the same result — the rule now called the “doctrine of equivalents.”

As the majority states, the Court indeed “construed” the “thing patented.” 56 U.S. at 338 . The “construction” was the legal rule that the claim could be infringed by an equivalent structure. Having corrected this error of law in the jury instructions, the Court did not then answer the factual question for itself, as now does the Federal Circuit. The Court remanded for retrial to the jury, on the correct instruction of law:

Whether, in point of fact, the defendant’s cars did copy the plaintiffs invention, in thé sense above explained, is a question for the jury, and the court below erred in not leaving that question to them upon the evidence in the case, which tended to prove the affirmative.

56 U.S. at 344 .

In Silsby v. Foote, 55 U.S. (14 How.) 218 , 14 L.Ed. 391 (1853), also relied on by the majority, the Court again did not remove factual issues from the jury. The Court construed the patent claim as a combination claim, and stated that the trial judge correctly instructed the jury on the law that all of the necessary parts of the claimed combination must be present in an infringing device. These were indeed matters of law. The Court stated that the trial judge properly left to the jury the question of which parts of the claimed device were necessary to its operation (which was to regulate the heat of a stove by automatically varying the position of the damper in response to temperature changes), as well as whether the defendants used these necessary parts.

The defendants had argued that the trial judge had impermissibly left a question of *1022 law to the jury. The Court pointed out that the question of which parts were necessary to regulate the heat of the stove was not a matter of law, but a question of fact to be decided by the jury:

The substance of the charge is, that the jury were instructed by the Judge, that the third claim in the specification was for a combination of such parts of the described mechanism as were necessary to regulate the heat of the stove; that the defendants had not infringed the patent, unless they had used all the parts embraced in the plaintiffs combination; and he left it to the jury to find what those parts were, and whether the defendants had used them.

We think this instruction was correct. The objection made to it is, that the court left to the jury what was matter of law. But an examination of this third claim, and of the defendants’ prayers for instruction, will show that the Judge left nothing but matter of fact to the jury. The construction of the claim was undoubtedly for the court. The court rightly construed it to be a claim for a combination of such of the described parts as were combined and arranged for the purpose of producing a particular effect, viz., to regulate the heat of a stove.

_ But the defendants also desired the Judge to instruct the jury that the index, the detaching process, and the pendulum, were constituent parts of this combination. How could the Judge know this as matter of law?

Id. 55 U.S. at 225-26 . The Court affirmed that the factual question of what the claim covered was for the jury to decide, in the course of determining the question of infringement. Indeed, the Court’s query was pointed: “How could the Judge know this as a matter of law?” Id. at 226 .

In Coupe v. Royer, 155 U.S. 565 , 15 S.Ct. 199 , 39 L.Ed. 263 (1895) the Court held that there was legal error in the trial judge’s description of the invention to the jury, and in the withdrawal of the question of infringement from the jury. The Court held that the trial judge had omitted a limitation contained in the claims of the patent {viz., that the orientation of the machine was vertical). It was indeed legal error to omit a claim limitation, then as now, and the Court, correcting this error, remanded for a new trial to the jury. The Court declined to give a peremptory instruction to the jury, stating that all of the differences are “the subject of legitimate consideration by the jury”:

[T]he question of infringement, arising upon a comparison of the Royer patent and the machine used by the defendants, should be submitted to the jury, with proper instructions as to the nature and scope of the plaintiffs’ patent as hereinbefore defined, and as to the character of the defendants’ machine.

155 U.S. at 579-80 , 15 S.Ct. at 205 . This case again illustrates the Court’s role as assuring that the law is correctly stated to the jury, and the jury’s role as trier of fact. Again, a case relied on by the majority does not support the majority’s position.

In Bischoff v. Wethered, 76 U.S. (9 Wall.) 812 , 19 L.Ed. 829 (1870) the Court distinguished between the construction of the patent as a legal instrument, and the factual nature of the thing invented:

It is not the construction of the instrument, but the character of the thing invented, which is sought in questions of identity and diversity of inventions.

76 U.S. at 816 . The issue was identity of invention, and the Court reiterated that the meaning of disputed terms of art is “a question of fact for the jury.” Id. at 814 . The majority includes Bischojf as authority for its removal of these findings of fact from the jury. That is a curious reading of the holding in Bischojf:

A case may sometimes be so clear that the court may feel no need of an expert to explain the terms of art or the descriptions contained in the respective patents, and may, therefore, feel authorized to leave the question of identity to the jury, under such general instructions as the nature of the documents seems to require. And in such plain cases the court would probably feel authorized to set aside a verdict unsatisfactory to itself, as against the weight of the evidence. But in all such cases the ques *1023 tion would still be treated as a question of fact for the jury, and not as a question of law for the court. And under this rule of practice, counsel would not have the right to require the court, as a matter of law, to pronounce upon the identity or diversity of the several inventions described in the patents produced.

Id. (emphasis added). Indeed, only two years later the Court again considered the issue, and in Tucker v. Spalding, 80 U.S. (13 Wall.) 453 , 20 L.Ed. 515 (1872) the Court held that a prior patent and related expert testimony on the issue of “diversity or identity” were improperly withheld from the jury, describing the issue as a “mixed question of law and fact,” and stating:

Whatever may be our personal opinions of the fitness of the jury as a tribunal to determine the diversity or identity in principle of mechanical instruments, it cannot be questioned that when the plaintiff, in the exercise of the option which the law gives him, brings his suit in the law in preference to the equity side ,of the court, that question must be submitted to the jury, if there is so much resemblance as raises the question at all. And though the principles by which the question must be decided may be very largely propositions of law, it still remains the essential nature of the jury trial that while the court may on this mixed question of law and fact, lay down to the jury the law which should govern them, so as to guide them to truth, and guard them against error, and may, if they disregard instructions, set aside their verdict, the ultimate response to the question must come from the jury.

80 U.S. at 455 .

In Winans v. New York and Erie R. Co., 62 U.S. (21 How.) 88 , 16 L.Ed. 68 (1859) infringement was conceded, and the issue at trial was “originality.” The Court stated that the trial judge “has given the only construction which the language of this specification will admit,” id., 62 U.S. at 101 , in explaining to the jury that the invention was in the manner of arranging the wheels and the car body. Having explained the invention to the jury, the question of originality was held to be for the jury, not the court.

In discussing the appropriate use of expert witnesses, the Winans Court stated that “professors or mechanics” can not prove “legal construction of any instrument of writing,” but may testify on matters of art or science:

Experts may be examined to explain terms of art, and the state of the art, at any given time. They may explain to the court and jury the machines, models, or drawings, exhibited. They may point out the difference or identity of the mechanical devices involved in their construction. The maxim of “unique in sua arte credendum” permits them to be examined to questions of art or science peculiar to their trade or profession; but professors or mechanics cannot be received to prove to the court or jury what is the proper or legal construction of any instrument of writing.

62 U.S. at 100-01 . On this aspect, too, the case does not stand for the removal of factual findings from the jury; indeed the Court recognized the various kinds of evidentiary facts on which technical experts routinely testify in patent cases.

In Heald v. Rice, 104 U.S. 737 , 26 L.Ed. 910 (1882) the Court stated that when there was no dispute about the technology, no need for evidence, and no question of fact requiring resolution by a jury, the “mere comparison” of a reissue and original patent was a matter of law for the court. Other cases related to a directed verdict when no fact was in dispute, e.g., Singer Mfg. Co. v. Cramer, 192 U.S. 265 , 24 S.Ct. 291 , 48 L.Ed. 437 (1904) (the trial court should have granted a directed verdict when there was no dispute as to the meaning of any term of art and no substantial evidence of infringement); or the grant of a new trial, e.g., Market St. Cable Ry. Co. v. Rowley, 155 U.S. 621 , 15 S.Ct. 224 , 39 L.Ed. 284 (1895) (since the facts were not disputed and no extrinsic evidence was given or needed, the court should have instructed the jury on lack of patentable novelty; the Court remanded with directions to set aside the verdict and grant a new trial). The new trial and the directed verdict are modes of judicial management of the trial process, and quite different from the majority’s decision simply to eliminate the jury. -

*1024 Hogg v. Emerson, 47 U.S. (6 How.) 437 , 12 L.Ed. 505 (1848), another case relied on by the majority, was part of a lengthy litigation. There was a jury trial, review by a circuit panel, retrial to a jury, and two appeals to the Court. In this appeal the Court considered which documents were properly considered when “construing” the patent, in view of the fire that destroyed the Patent Office files in 1836. In reviewing the question of whether the patent in suit covered the entire steam engine or only the improvement, the Court “construed” the patent as covering only the improvement. Id., 47 U.S. at 484 . The Court affirmed the trial court, which the report states “left the question of fact as to reasonable diligence of the patentee or not in this respect, and also all questions of fact involved in the points of the case for the defendants, to the jury.” Id. at 445 .

The majority also relies on Levy v. Gadsby, 7 U.S. (3 Crunch) 180, 2 L.Ed. 404 (1805), wherein the Court ruled that a certain document was a contract and not some other form of transaction, and was subject to the usury law. This was legal construction of a document, and was decided by the Court. This case says nothing about removing disputed factual questions from the jury. The majority also cites Eddy v. Prudence Bonds Corp., 165 F.2d 157, 163 (2d Cir.1947), cert. denied, 333 U.S. 845 , 68 S.Ct. 664 , 92 L.Ed. 1128 (1948), wherein the court, reviewing the legal operation of a court-approved Supplemental Trust Agreement in bankruptcy in view of a court' order, stated that “appellate courts have untrammeled power to interpret written documents.” This determination of legal effect is indeed “construction” of a legal document.

I shall not dwell on the majority’s reliance on other cases that were bills of equity and tried to the court, for they do not raise the issue of the jury right. See Loom Co. v. Higgins, 105 U.S. 580 , 26 L.Ed. 1177 (1881) (determining if patentee was the true inventor, and whether patent claim was sufficiently described in the specification); Goodyear Dental Vulcanite Co. v. Davis, 102 U.S. 222 , 26 L.Ed. 149 (1880) (limiting Goodyear’s claims to dentures manufactured by vulcanization); Bates v. Coe, 98 U.S. 31 , 25 L.Ed. 68 (1878) (determining which elements constituted the invention and which constituted equivalents); Merrill v. Yeomans, 94 U.S. 568 , 24 L.Ed. 235 (1876) (construing patent claim in light of specification as only for process of manufacture); and Seymour v. Osborne, 78 U.S. (11 Wall.) 516, 546 , 20 L.Ed. 33 (1871) (discussing “scientific witnesses to aid the court in coming to a correct conclusion”). The majority cites Merrill v. Yeo-mans as “applying ‘well-settled rules of construing all instruments.’ ” The rule the Court applies at this quotation is that words and phrases are to be construed so as to give them meaning. 94 U.S. at 571 . This is indeed a rule of law; the Court did not convert findings of fact into rules of law. Nor should it be necessary to point out that when cases are tried “to the court,” majority op. at 996-997, the resolution of disputes as to what claim terms mean is indeed “for the court.”

Exhibit Supply Co. v. Ace Patents Corp., 315 U.S. 126 , 62 S.Ct. 513 , 86 L.Ed. 736 , 52 USPQ 275 (1942), also cited by the majority, turned on prosecution history estoppel resulting from an amendment to the claims in the Patent Office. The Court stated the rule of law that “what the patentee, by a strict construction of the claim, has disclaimed ... cannot now be regained by recourse to the doctrine of equivalents, which at most operates, by liberal construction.” There was no dispute as to the meaning of technical terms, and the Court applied this rule of law to the undisputed facts. This had been a bench trial, on bill of equity. Ace Patents Corp. v. Exhibit Supply Co., 119 F.2d 349 , 48 USPQ 667 (7th Cir.1941). It is difficult to discern the relevance of this case to the issues in Markman.

Many dozens of patent cases reached the Supreme Court. Some of those relied on by the majority as support for trial to the court were bills in equity. Of those in law, most were tried to a jury. It is not possible to diminish the great weight of precedent wherein patent infringement was tried to a jury, the jury deciding disputed factual questions of what the patent covered, and applying these findings to the accused device. The court today effects a dramatic realign *1025 ment of jury, judge, and the appellate process.

D. THE SPECIAL RESPONSIBILITY OF THE FEDERAL CIRCUIT

The Federal Circuit is responsible for establishing consistent national law in its areas of assigned subject matter. The court early in its existence took note that patent cases were only one of many areas of commercial dispute, only one of many areas of intellectual property dispute, that are tried in the district courts. We have striven to assure that unnecessary burdens are not placed upon the district courts of the nation by virtue of the separate path of appellate review of patent cases. We acted to assure that the same procedures would apply in the trial of patent cases as in other civil actions. See, e.g., Allen Organ Co. v. Kimball Int’l, Inc., 839 F.2d 1556, 1563 , 5 USPQ2d 1769, 1774 (Fed.Cir.) (for matters not unique to patent law the procedural law of the regional circuit applies in patent trials), cert. denied, 488 U.S. 850 , 109 S.Ct. 132 , 102 L.Ed.2d 104 (1988). Thus the litigation process that served other civil disputes also served in patent litigation. Today’s ruling, with its departures from the rules of evidence, its changed standards of deference and review, its conflict with established jury and ‘bench procedures, challenges the principle on which this comity was based.

Patent cases are not unique in their usage of specialized terms and words of art, in their reliance on technologic or scientific evidence, in their dependence on findings of technologic fact. Evidentiary conflicts with respect to technology and science arise in a variety of cases; and the conflicting testimony of expert witnesses is ubiquitous. Trial judges have extensive experience in assuring a fair trial, and finding, within human limitations, the truth. 12 Today this court severs patent cases from all others, requiring different (and uncertain) procedures at trial, taking unto ourselves a different, and uncertain, appellate role.

It is the responsibility of the appellate court to assure that the law is correctly stated. The rules of patent law include an ever-enlarging body of nuance and clarification, flowing from twelve years of Federal Circuit jurisprudence and the rich history on which we have built. This court has undertaken the fine-tuning of the law, appropriate to the importance of technology in today’s world. Much of this fine-tuning relates to new fields of science and technology (computers, biotechnology, materials); but it also relates to traditional concepts of patent law as applied to modern technologic and commercial needs.

The appellate role is to apply these principles in wise implementation of the policy of the law, as litigants probe the grey areas that test conflicting policy considerations. The appeal is not designed for de novo finding of the facts. I doubt the practical féasi-bility of the majority’s holding that this court will “construe” the meaning of technical terms and words of art without benefit of the trial experience. It is of course appropriate for this court to be alert to methodologies of resolution of disputes that involve science and technology. The trial of scientific/tech-nologic disputes was explored, for example, in the Report of the Carnegie Commission, Science and Technology in Judicial Decision Making (1993); the Report of the Brookings Institution, Charting a Future for the Civil Jury System (1992); and in ongoing studies and Reports of the Federal Judicial Center. However, in this complexity of problems and solutions, it is an illusion to think that patent litigation difficulties can be resolved by turning factual issues into matters of law and assigning them to the Federal Circuit.

The deference that appellate courts must give to the trial process is fundamental to the *1026 efficiency, and the effectiveness, of the judicial system. It implements the two-tier litigation right, and provides stability to the trial process while preserving appellate authority for the law, its policy and its purposes. The court’s decision today denies the critical values of the trial, and moves the Federal Circuit firmly out of the juridical mainstream.

y

THE MERITS

Both sides testified on the meaning and scope of the term “inventory,” as used by Markman and in light of the Westview system. The issue was whether “inventory” meant only clothing, or could reach the invoices of the Westview system. Markman presented four witnesses. Westview presented one witness. After the jury verdict in favor of Markman the district court, applying recent Federal Circuit panel opinions that required de novo determination of the issue (foreshadowing today’s en banc holding), reviewed the evidence independently and decided in favor of Westview. The district court did not discuss the jury verdict, or state whether there was evidentiary support for the jury verdict.

The district court did not apply the proper standard on post-trial motions, viz. whether there was substantial credible evidence of such quality and weight that a reasonable jury could have reached the verdict that was reached by this jury. It is for the trial judge to decide, in the first instance, whether the jury verdict can stand, or whether the judgment should have been directed, or whether a new trial should be granted. I would remand for redetermination on the correct standard.

. The Federal Circuit, finding this fact en banc, holds that the inventor Markman’s testimony is "of little or no probative weight” to explain his invention, apparently because he was represented by an attorney before the patent office. The majority states that it is “not unusual” for the inventor not to know what his attorney has patented. Maj. op. at 986. This will be a revelation to the nation's patentees. The majority earlier in its opinion "rejects” Markman's testimony, maj. op. at 983, apparently based on its weight, a question of fact, not on its admissibility, a ruling of law.

. In Graham v. John Deere the Court described the issue of "obviousness” in patent cases as one of law based on underlying facts. An analogous pattern has heretofore applied in connection with "claim construction”, i.e., as a question of law based on underlying facts. I suggest that these ultimate questions have a strong policy component, and that the Federal Circuit's responsibility for imparting consistency to patent decisions is a significant factor in the law/fact dichotomy.

. Nathan Isaacs, The Law and the Facts, 22 Co-lum.L.Rev. 1, 13 (1922) (warning courts of "medieval assumptions as to the omniscience of the learned man”).

. English patent cases were cited by United States courts well into the nineteenth century.

. Attempting to understand how this procedure would work for complex technologies, at the en banc argument I inquired of Westview's counsel:

J. Newman: If the claim is sufficiently complex — technologically or just complex in general — that the judge can't decide what it means without taking testimony, hearing the experts, hearing the inventor, hearing whatever else it is that each side needs to, wants to, adduce that the judge permits ... I can envision, can't you, that a judge would have to hold some kind of evidentiary hearing, at least, if not a mini-trial, in order to learn enough about the claim to decide, as a matter of law, disputed issues?

Mr. Griffin: Yes your honor.

J. Newman: We are assuming the issues are disputed, that this is not just a matter of explanation, but a matter of requiring a choice between one side's viewpoint and another. And that the judge should then have a preliminary trial to decide what the claim means by making whatever choices need to be made, and then tell the jury: "Take it from here now, apply this to the accused device”?

Mr. Griffin: Yes your honor.

J. Newman: That is quite unusual, is it not? Have you seen this done?

Mr. Griffin: Personally no your honor.

J. Newman: I wonder how the trial judges would take to that.

Mr. Griffin: Probably would not like it your honor, because that would impose a burden which in most cases can be avoided.

. Among the rare exceptions to deferential appellate review of factual findings are the “constitutional facts”, discussed in Bose Corp. v. Consumers Union of United States, Inc., 466 U.S. 485 , 104 S.Ct. 1949 , 80 L.Ed.2d 502 (1984). In Bose the Court cautioned against enlarging its holding *1010 beyond the conflict between constitutional provisions there exemplified. The “constitutional fact” exemplified in Bose does not place all facts in the hands of appellate courts for de novo finding. Such exceptions to the otherwise firm rule of deference to the trier of fact have always been narrow. See generally Frank R. Strong, Dilemmic Aspects of the Doctrine of "Constitutional Fact", 47 N.C.L.Rev. 311 (1969).

. The date of the Statute of Monopolies is variously reported as 1623 or 1624 depending on *1012 whether the old or new English calendar is used. Edward C. Walterscheid, The Early Evolution of the United States Patent Law: Antecedents (Part 2), 76 J.Pat. & Trademark Off.Soc'y 849, 873 n. 98 (1994).

. There were two courts in Chancery, the Ordinary court and the Extraordinary court. The Extraordinary side of Chancery was so termed because matters requiring the exercise of the King’s Conscience were there addressed, for extraordinary relief. Abridgment at 127; Coke, 4 Inst, at 79. The Ordinary side of Chancery has been referred to as “common-law chancery,” see Kirker v. Owings, 98 F. 499, 506 (6th Cir.1899), and proceeded according to the laws and statutes of England, exercising the ordinary powers of Chancery. Middle Temple, General Abridgment of Cases in Equity 127 (1739) ("Abridgment”); Coke, 4 Inst, at 79.

The Ordinary court had the power to repeal letters patent, by plea of scire facias. However, if the matter "descended to issue" the court was without jurisdiction to try it to a jury, and the Chancellor would direct the issue to a court of law, where the issue would be tried to a jury, “because for that Purpose both Courts are but one.” Abridgment at 128; see Coke, 4 Inst, at 80. After trial, with jury verdict rendered, the cause was returned to Chancery for further disposition consistent with the verdict. Abridgment at 130 ("A Cause shall not be examined upon Equity in the Court of Requests, Chancery, or other Court of Equity, after Judgment at the Common Law.”)

. Carpmael lists the date of the decision as 1758, and Abbott, supra, at 9, states the date as 1766. The date of the patent grant is 1758, suggesting that Abbott may be correct. There is no official report of Dolland’s Case, but the decision is discussed in Boulton v. Bull, 2 H.B1. 463, 482-87 (K.B.1795).

. Hanbury, supra, at 87-88, explains that the court was not instructing the jury on what verdict it must render. The practice was for the judge to summarize the evidence and the testimony, and frequently provide a “hint” to the jury. However, this did not diminish the authority of the jury to decide the matter.

. Although some amici curiae encouraged the Federal Circuit to find technological facts for ourselves, none explained the procedure by which we are to do so. Are we to read the entire record of the trial, re-create the demonstrations, decipher the literature of the science and art; are we to seek our own expert advice; must the parties be told the technical training of our law clerks and staff attorneys? No amicus explained how improved technological correctness — that is, truth — would he more likely to be achieved during the appellate process of page-limited briefs and fifteen minutes per side of argument.

. Many aids to the trial process are at hand when the issues are scientific and technologic. The Manual for Complex Litigation, Second (1985) points out the utility of special verdicts and interrogatories, see § 21.633, and that Fed. R.Evid. 706 is particularly useful when experts have divergent opinions, see § 21.51. Important studies have been made, see the Federal Judicial Center's Reference Manual of Scientific Evidence (1994). No study that I have seen or heard of proposes simply to turn complex factual determinations of technical issues over to the appellate court. To replace the trier of fact with the Federal Circuit is as unfriendly to the search for truth, as it is unworkable.